An Inaugural Lecture Delivered on 13 October 1993

Fallacies of Competition:
Myths and Maladjustment in the Third World
John Weeks
Professor of Development Economics in the University of London And Director Centre for Development Policy and Research

[Forthcoming in Spanish 2009]

The Competitive Hypothesis

For decades the world economy was been racked by severe instability. Compared to the twenty years from 1953 (end of the Korean War) to 1973 (first oil price increases), the subsequent decades have been characterized by turmoil on international markets, which in the developed countries manifested itself in financial instability and recessions. For the developing countries, especially Africa south of the Sahara and Latin America, circumstances were considerably grimmer, prompting some to refer to the 1980s as the Lost Development Decade, with the subsequent years ones of slow growth for most countries. The economic collapse of the 1930s called forth a radically new treatment of macroeconomic policy, associated with but certainly not limited to J. M. Keynes (and we have with us tonight one of the pioneers of that new paradigm, Hans Singer). In contrast, the instability of the last twenty years has seen a reversion in theory and policy to the doctrines of the ancien regime of economics, the paradigm of automatically adjusting full employment, what is called neoclassical economic theory, in its purest and most free- market form. For the problems of the underdeveloped countries, falling per capita income, inflation, balance of payments crises, and unsustainable fiscal deficits, the neoclassicals have a story to tell. The Bretton Woods organisations have been the vehicles to transform those stories into practice. The story goes as follows: through misguided policies based upon state intervention, governments of developing countries have mismanaged their economies to the point of disaster and beyond. These mismanagements all share the common characteristic that they restrict competition. Attempts to foster national development goals by restricting import access to domestic markets (protectionism) resulted in distorted internal prices that discouraged exports and generated inefficient production. Expansionary

macroeconomic policies designed to keep national output near its potential resulted in inflation. The accumulation of debt represented a major aspect of the imprudent state behaviour. The inflation fed back into the economy, resulting in further distortions, most notably of the exchange rate, making exports too expensive and imports too cheap. In this context, it is small wonder that the investment climate deteriorated and the private sector withered.

The story provides the solution to this woeful collection of self- inflicted ills: restriction of government action. Most importantly, trade liberalisation is necessary to set the economy right. This would allow world markets to signal to private producers what is profitable and what is not.1 To facilitate the private sector to get on with the job of setting right what governments have irresponsibly mismanaged, the state sector must be reduced, which will have the benign side-effect of reducing expenditure and budget deficits. The intended effect of these measures is to increase competition, because it creates the best of all possible economic outcomes. Competition generates economic efficiency, and, to quote from the Penguin Dictionary of Economics: Economic efficiency [refers to] [t]he state of an economy in which no one can be made better off without someone being made worse off. [T]hree types of efficiency hold. The first is productive efficiency, in which the output of the economy is being produced at the lowest cost. The second is allocative efficiency, in which resources are being allocated to the production of the goods and services the society requires. The third is distributional efficiency, in which output is distributed in such a way that consumers would not wish...to spend [their] incomes in any different way.2 In summary, the neoclassical proposition is that underdeveloped countries have suffered from the self-inflicted wounds of government regulation, which have denied societies the benefit of competition. Eliminate the interventions and competition will result, bringing a situation of economic efficiency, in which no improvement is possible without making some people worse off. This lecture will address that analytical conclusion.

Fairy Tale as Hypothesis

Some of you may be of the generation that can recall an American song of the 1950s, Young at Heart, which begins: Fairy-tales can come true, they can happen to you... These lines might represent the theme song of the neoclassical proposition. The comparison to fairy tales is appropriate when one recalls that a large number of these tales end with the expectation, they all lived happily ever after. This state of

grace in fairy tales is usually achieved through the ex machina intervention of magic. Magic is normally used to dispatch trolls or dissolve witches. In neoclassical economics the trolls or witches in question are officials of the state, the elimination of whom results in a specific state of grace called Pareto Optimality, known more colloquially as perfect competition. We find an example of economic munchkins living happily ever after in trade theory, in which a welfare-constrained population is rescued from the evil kingdom of autarky and delivered unto the bliss of free commerce via the gains from trade. This delivery is all the more wondrous because the last shall be first: it is the small countries of the world that by the magical means of trade liberalisation gain most, while the mighty gain least. The full wonder of this fairy tale derives from the ease by which it is converted from fable to fact: it can be achieved merely by agents instructing their governments to do nothing; to stop all the myriad of irrational actions that stand between sub-optimal reality and the ideal of free competition.

Searching for Competition

On the basis of fabulous neoclassical parables, governments of developing countries (and developed) are entreated to reduce labour market regulations, grant sweeping powers to central banks, and eliminate trade restrictions. One would presume that the theoretical underpinnings for such policy recommendations are powerful, indeed. The key element in the theoretical edifice is the concept of competition. At the level of the high theory of Nobel Laureates and discussions on the street, competition is presented as the vehicle by which the benefits of capitalism spread though the land. Competition ensures that consumers receive value for money, competition forces producers to lower costs, and generates the ultimate benefit of the market system, choice. In the absence of competition benefits fade and markets whither. Competition, more than just a good thing, is the Philosophers Stone of neoclassical economics. Touch it to a market and efficiency prevails. When competition holds sway in the neo-classical sense, the working of the economy approaches the sublime; when it is imperfect all necessary steps must be taken to purify it. Each action of government must be fiercely scrutinised for its probable negative effects on

competition. In the profession and in the press, the truth of these arguments is taken as self-evident. The deconstruction of neoclassical fables must begin with competition. Let us inspect what the neoclassicals mean by the term. Again, the Penguin Dictionary of Economics, which states: competition, see perfect competition (the same advice is found in other dictionaries). This cross-reference must produce disquiet, for it implies that the definition will be narrow, in terms of an ideal rather than a general treatment to cover competition in practice as well as the abstract. The disquiet is confirmed: [Perfect competition is] a model of industrial structure in which many small firms compete in the supply of a single product. [Four] primary features characterize a perfectly competitive industry: (i) There is a multitude of firms all too small to have any individual impact on market price. (ii) All firms aim to maximize profit. (iii) Firms can without cost enter and exit the industry. (iv) Outputs are homogeneous.3 The authors of the dictionary anticipate that the reader might not immediately recognize the importance of such an improbable concept: Although the features of perfect competition make it look a poor description of modern industry, it is a realistic description of world commodity markets where many traders deal in a homogeneous product. One must lament that there are very few world commodity markets with homogeneous products. Anticipating this insight on the part of the reader, the entry continues: [The] very powerful results [of the theory of perfect competition] indicate that the achievement of even a partially competitive market can be advantageous. We should read this statement in light of the entry on economic efficiency previously quoted. At the end of that entry we find the statement: Economic efficiency will exist in an economy in which perfect competition characterizes every sector. Therefore, the statement that perfect competition indicates the achievement of even a partially competitive market is the equivalent of saying that imperfect markets approximate the benefits of economic efficiency. It is part of the current folk lore of economics that perfect competition is best, but if that cannot be achieved, more competition is better than less. Almost fifty pears ago Lipsey and Lancaster demonstrated that this proposition is invalid. As surprising

as it may seem, mainstream economic theory provides no basis for systematically concluding that more competition is better than less: perfection is perfect and no conclusion can be reached about varying degrees of impurity. In the current

ideological mood of the profession, this insight is singularly unwelcome. The agnostic conclusion about degrees of competitive impurity reflects the lack of an analysis of competition as process. Most textbooks state that a large number of small buyers and sellers insures that each acts on the belief that he/she cannot affect the market price. This common statement is a logical muddle: the number and size of firms are characteristics; that the firms may or may not be able to impact on market price is an outcome. The two need to be linked by a process. It may seem common sense that in such a market buyers and sellers would believe themselves unable to affect price, but even commonsensical jumps in logic are not permitted in abstract theory. The issue of the nature of the competitive process is better illuminated in the Pan Dictionary of Economics and Commerce, which defines competition bereft of modifiers: [Competition is a] situation in a market in which a number of producers are attempting to increase their own profits at the expense of rival producers. This leads to price wars, attempts to increase market shares, product differentiation, etc.4 This definition of competition conforms to what one observes. After this promising beginning, the entry goes on to say: Competition is most keen under conditions of perfect competition... This cannot be correct: recall that perfect

competition is defined as a situation in which no producer can impact on price. Therefore, producers in perfectly competitive markets are excluded by definition from price wars and struggling for market shares. They are also explicitly prohibited from product differentiation. The characteristics of a market that provide for competition in its most perfect neoclassical manifestation exclude the forms which competition might rake.5 Therefore, it is difficult to divine how the analysis of perfect

competition offers insights into less than perfectly competitive markets, when price and product characteristics are an instruments of rivalry. Unicorns come to mind in this context: one can define such a beast as a four-legged horse-like creature with a single horn and elaborate its anatomical characteristics in some detail; but definition

does not imply existence. Some might argue along neoclassical lines that study of the unicorn provides powerful insights into the nature of horses. I have to an extent been attacking a straw-person, for the cited definition of competition (many buyers and sellers, homogeneous product, etc.), encountered as commonly as it is in textbooks, is low and vulgar theory. The high-theory treatment of competition is found in Walrasian general equilibrium analysis. No serious neoclassical discussion of competition is possible outside a general equilibrium framework. Before entering into this rarefied region, we should review why it is necessary. The intent was to assess the economic programmes of liberalisation widely implemented in developing countries at the encouragement of multilateral and bilateral lenders. These programmes are justified on the argument that competition produces social benefits. I sought a definition of competition, only to discover that the low-theory version proved an analytical dead-end. Thus, we are driven to the realm of the Nobel Laureates to answer the question: why and under what circumstances is competition a good thing? Competition is supposed to foster efficiency; this increased efficiency is supposed to bring welfare gains to the population. Efficiency is achieved by the interaction of buyers and sellers generating prices which feed back upon those buyers and sellers resulting in an allocation of resources which is optimally efficient. Prices are the key: competition is good because it produces prices which determine socially efficient decisions by producers and consumers. Governments distort these efficient prices when their actions result in any combination of prices which is other than that generated by buyers and sellers exchanging in the absence of government action.6 Under these circumstances, buyers and sellers are said to be price-constrained. The decisions they take are based upon the prices they observe in the market. The process of competition (as yet unexplained) is a good thing because it produces a set of prices which generates the best conceivable allocation of resources. If buyers and sellers make their decisions on the basis of market prices, then prices are beyond their control. Were this not the case, buyers and sellers (agents) would not be

constrained by prices; they would be making decisions about prices, not on the basis of prices. It follows that these agents are not constrained by their incomes. Buyers choose their incomes just as they choose how much beer, shoes, etc. to buy;

analogously, firms can chose any level of output, no matter how large or small, and select the one that gives them their greatest profit. It may seem strange that firms would not worry about how much they can sell and people would not worry about obtaining the amount of paid work they desire. Yet these propositions are inseparable from the argument that competition produces efficient and desirable outcomes. They are ultimately the theoretical justification of IMF/WB stabilisation and structural adjustment programmes; they provide the general argument that makes it unnecessary for these institutions to justify their policies on a country-by-country empirical basis. Keynes pointed Out that these assumptions represent a special case. Consider an economy with ten per cent unemployment of labour and idle industrial capacity. In such circumstances it is reasonable to presume that the idle capacity exists because firms discover that they cannot profitably sell more output; the unemployed have looked for work and found none available. The prices that prevail in such a situation are not vehicles of allocative efficiency. They cannot possibly be, because some labour and some capital are idle: resources are not scarce. In this case, prices reflect that the economy is income or demand constrained. Were demand to increase, outputs would rise and prices would change: some prices would go up and others would go down, generating a different allocation of resources. Only when all resources are active will it be the case that prices assume the autonomous, allocative role that they play in the parable of competition. To return to our central theme, the deregulation of markets has the purpose of fostering competition. In order for that policy to be generally defensible, competitive markets should produce an outcome in terms of relative prices which is efficient and which would not be achieved were markets regulated. If there are idle resources, the argument is a non-starter. The prices which prevail in the market would not be determined by competition among buyers and sellers over scarce resources, but by levels of demand. The white magic of competition cannot be justified on a market-by-market basis (partial equilibrium in the jargon), but delivers fabulous benefits only on the grander scale of global full employment of resources (full employment general equilibrium). In the absence of full employment of resources, it cannot be

demonstrated theoretically that any specific competition-fostering measure will lead to increased efficiency (improved allocation of resources) and welfare gains to the population. 8

To work its magic, competition must be the vehicle for generating full utilisation of resources for all markets taken together (the economy as a whole). A global competitive full employment is the necessary condition for the efficient operation of each specific market. The achievement of general equilibrium is logically prior to establishing competitive outcomes in partial equilibrium. An example of this is international trade theory, which assumes full employment as the basis for initiating the analysis of the gains form trade. Rare is the trade theorist who explores the implications of unutilised resources.7 After much wandering, we discover that to answer the question, why is competition a good thing, we are led necessarily to the question, how does competition generate full employment of resources?

The Willing Suspension of Disbelief

Coleridge wrote that the pleasure from fiction derives from a willing suspension of disbelief Neoclassical economics treats such self- induced credibility as method. In demonstrating how a competitive system produces a full employment general equilibrium the principle of Coleridge achieves its fullest expression The theory of general equilibrium was invented by L. Walras at the end of the nineteenth century. In a Walrasian world, people come to the market each with a fixed amount of commodities to sell and the purpose of making exchanges to acquire a different set of commodities. This different set derives from their personal preferences for each commodity determined prior to the opening of the market. The traders compare the prices of what each has to sell with the prices of what is to be bought and chose the most advantageous combination of buying and selling. At first glance, this approach does not appear very promising for the purpose of establishing competitive full employment since no production is involved Commodities come to the market already produced, with the only important action to unfold being the determination of a set of prices that will ensure that there are no surpluses or shortages (called excess demands and excess supplies). Even the process of price formation in this world of fixed supplies is not very satisfactory. If traders are left to their own devices, the probable result will be a set of prices that leaves some commodities unsold. This would occur, for example, if on the basis of an initial set of prices the seller of beer received a price so low that that

he/she was unable to make the required contribution to the total expenditure that would allow all apples to be sold. One way out of this problem would be to allow for commodities to be sold at different prices during the market period. This is what happens in real markets: on the sell it or smell it principle, left-over apples would be dumped at a below-cost price just as the market closes. This cannot be allowed if the trading process is to produce an efficient outcome. If any commodity sells at more than one price during the market period, this is prima facia evidence that trading was demand constrained; i.e., sales were not determined by relative prices, but relative prices were established by the need to sell certain quantities. The Walrasian process is designed to avoid this. To avoid it, Walras introduced an ex machina auctioneer. The role of the auctioneer is to stand at the centre of all traders and listen to the alternative offers by buyers and sellers. He/she is invested with the draconian power to prohibit any trades at prices that leave excess demands and excess supplies, which we shall call disequilibrium prices. No trading at disequilibrium prices is permitted. The

auctioneer is wise indeed; he/she must prevent the manifestation of disequilibrium; therefore, the auctioneer must have perfect foresight in which he/she knows in advance the trading outcome of every set of prices, even though all save one cannot be observed. One should imagine a feckless trader making a purchase order, only to have the auctioneer boom out, Not so fast, Jones, that offer for two jars of mustard will leave eight bottles of Budweiser unsold in Chicago. Actual markets do not have auctioneers except in very special cases; and real auctioneers could hardly care less about the general equilibrium purity of the prices they gavel down. Markets, with or without auctioneers, do not clear simultaneously, but sequentially; there is no going back if the wrong price is established for Marmite. Nothing remotely resembling a Walrasian market exists in any exchange economy, yet such markets are taken as the basis of neoclassical competitive theory. It is an interesting sociological phenomenon that such a patently absurd view of market operation should be incorporated into a social science. More interesting still, this theatre of the absurd is treated as the norm and what actually occurs as a deviation from that norm that must be justified. Inversion of fact and fantasy is enshrined in official terminology. Following the proposal of Sir John Hicks in the 1930s, exchanges that result in excess demands and excess supplies are called false trading, with the implication that what happens in the imaginary market is true. Here one has 10

entered into a quasi-religious realm, in which the observed world is false, and the world of the imagination is true. After much tedious textual deconstruction, we discover the meaning of perfect competition. Many buyers and sellers will be unable to affect prices when they trade in a market with an omniscient auctioneer and no false trading is allowed. Competition is a considerably more problematical concept than the person on the street (or most students of economics) is led to believe. The problematical nature of neoclassical competition becomes all the worse when one attempts to include within the Walrasian framework the most important traded commodity, labour power (labour services is the term neoclassicals prefer). It is the clearing of the labour market, above all others, that determines full utilisation of resources in a capitalist economy. To include it as one of the many traded items in the Walrasian market requires the introduction of production into the system. Production creates a myriad of difficulties, the analysis of which lies beyond the scope of this presentation. It is sufficient to point out that the market for labour power cannot be treated like the markets for produced commodities and services without flights of fantasy that make the mythical auctioneer seem a credible construction. In a Walrasian world workers arrive on the market day with their labour power to sell (labour services is the neoclassical term), and a price for it is struck such that every seller of labour power is content with the amount of time contracted for work, and every employer is content with the contracted labour time. On this basis, workers determine their incomes by choosing the optimal amount of work in light of wage rate offers and their preferences between income and leisure. Under these circumstances, workers, too, are price constrained, treating employment opportunities as unlimited at the prevailing wage Even in the abstract problems arise with this approach to the labour market. It would be somewhat credible to argue that at the outset of the potato market period no potatoes have been sold: the market for potatoes takes the form of a great collection of unsold potatoes.8 Such is not the case for the labour market. An excess supply of labour occurs when the vast majority of workers is employed. While the excess supply of any non labour commodity can reasonably imply disappointment on the

part of the vast majority of sellers, an excess supply of labour power is consistent with contentment for the vast majority of sellers. Further, the neoclassical equilibrating adjustment which would eliminate the excess supply, a lower real wage, would leave

11

the vast majority of sellers (the employed) worse off. This contrasts with the situation of the seller of a non-labour commodity, who loses from the fall in price, but gains from the rise in quantity sold. At any moment the vast majority of workers is not on the market. Thus, if wages fall or rise for the newly-employed, this is prima facia false trading, not evidence of allocative efficiency. We can conclude the following: neoclassical theory has no adequate explanation for marketing clearing which bears the most remote resemblance to the operation of labour markets. The theory provides no analytical basis for concluding that exchange economies tend automatically to full employment. Therefore, the theory provides no justification for treating economies as price constrained. Therefore, it cannot be established in the abstract that prices generate an allocation of resources that is economically efficient. In consequence of the foregoing, one reaches the conclusion that there is no theoretical support for the view that competition brings welfare and efficiency benefits, except in the Walrasian fairy tale. The theoretical arguments for a myriad of liberalisation policies, privatisation, tariff and quota reduction, internal markets for public services, ad infinitum, are built upon a foundation of sand. In the realm of political economy, Walrasian markets serve as a superficial justification for the ideology that markets function smoothly and harmoniously. The harmonious result is achieved through a fiction that by

assumption eliminates the forms competition can take and excludes a meaningful consideration of labour markets. In neoclassical theory competition is not an

analytical category (even less an empirical one), but rather the code word for the outcome it is assigned to produce. Neoclassical competition is supposed to be the mechanism that produces in practice the sublime state of economic efficiency. We discover that it is indistinguishable analytically from the result; indeed, it is the result itself. In the abstract one can, in flagrant violation of all experience, establish a state of bliss in which all resources are fully and optimally used. What is the real process that produces this result? No one knows. What is the theoretical process, however implausible, that produces this result? In one hundred years no economist has improved on Walras improbable answer.

Reconsidering Competition

12

Competition, the central concept in mainstream economics, has no content with regard to process. While neoclassical theory has no theoretical or empirical insight into competition, buyers and sellers do compete. To treat the process of competition we require a methodological break, which abandons the ex machina invoking of competition as a magic wand that turns the anarchy of the market into harmonious optimality. A number of great figures in economics have made the methodological break. A common characteristic of alternative approaches is that they allow for competition to generate destructive and chaotic outcomes; i.e., competition is not constrained by definition to generate a harmonious equilibrium in which all gain. Marx argued that competition is the process that undermines the accumulation of capital itself, by the destruction of fixed means of production through obsolescence. At the other ideological wing of the profession, Josef Schumpeter called competition the process of creative destruction, in which the bold and strong wipe out the weak and timid. Singer, along with Prebisch, argued that international trade involves competitive processes that systematically shift resources from peripheral countries (mainly but not exclusively underdeveloped countries) to centre countries (exclusively developed countries). The work of Singer is especially important. His conclusion that there exists a tendency for the terms of trade to move against peripheral countries is well-known. Less appreciated is the theoretical insight that this tendency results not only from uncompetitive markets in developed countries, but from the interplay of uncompetitive markets on the one hand and competitive ones on the other. Resource transfers via the terms of trade are as much the result of competition as they are of market power. With the decline of the Classical School of Political Economy, Smith, Ricardo, Marx and even Mill, the attempt to analyze competition disappeared from the mainstream. Only in a few isolated writers, Schumpeter and Singer already mentioned, Joan Robinson and E. J. Chamberlin in the 1930s, Myrdal with his theory of cumulative causation, and more recently Ben Fine, Laurence Harris, and Anwar Shaikh, does one find a rejuvenation of the classical insights. Reconstruction of the lost analysis of competition involves application of three basic principles: first, that competition be defined as the movement of capital; second, the integration of technical change with the movement of capital; and third, recognition that within industries the efficiency of production is unevenly developed. Each of these principles 13

breaks with the main stream. While the mainstream approach certainly refers to the mobility of capital it confuses capital with machinery and thus divorces it from labour. Wedding competition with the movement of capital implies that we cannot formulate a general analysis of competition equally applicable to markets dominated by capitalist enterprises and peasant producers, a point made by Terry Byres in his work on India and Elizabeth Dore in her treatment of grain markets in Nicaragua. Central to the neoclassical method is the treatment of technical change as a long-run phenomenon, which, in addition to confusing process with outcome, banishes technical change from the analysis of competition. Finally, by presuming all producers in an industry to be identical (the representative firm assumption of Alfred Marshall), all forms of conflict among them are eliminated. The short-run analysis of mainstream theory renders competition an empty category; in its long run analysis it allows for the form of competition, inter-industry transfer of resources, but renders it trivial by expelling the effects of technical change. To take an analogy, in a neoclassical-Walrasian Wimbledon, all contestants would be of equal ability, delivering their ground strokes along the same well-behaved production function; or if not, the inferior players would quickly learn from the superior and achieve equality, and in the absence of an external shock (such as an injury), no player would ever be eliminated and all matches would be add-in, add-out, or deuce, world without end. If one accepts that, in general, production units within industries vary in unit costs, then it follows that the movement of capital does not reproduce the average production conditions in an industry, but typically seeks to emulate or surpass the most efficient operator For the most part the movement of capital does not involve the physical shift of existing means of production but the construction of new production facilities The construction of new facilities is the realisation of technical changes Thus the movement of capital is a disruptive process far from establishing a harmonious equilibrium its purpose is to disrupt, to eliminate the weak and challenge the strong to force upon the industry a new standard of efficiency and cost Here Marxs insight is powerful the movement of capital to equalise profits across industries is the process of generating uneven development equilibration in exchange (equalised profit rates) hides the generation of uneven development in production. This civilised warfare of production and exchange (to use Marx s term) imparts to capitalism its dynamism It also involves major social costs the most 14

important being poverty unemployment and environmental damage When these costs manifest themselves, they are treated as incidental to the operation of markets with their costs forced upon employees the unemployed and small producers.

Competition and Adjustment

In the late 1980s, John Williamson of the Institute for International Economics published an article on structural adjustment in Latin America with the selfexplanatory title, What Washington Means by Policy Reform. In the article he summarises the policy reforms that the Washington Consensus viewed as constituting the path to vigorous and sustained growth.9 The same elements would apply to countries in Africa and Asia. The consensus in question is between the international financial community (including as point-persons the IMF and the World Bank) and the governments of the major industrial countries. The extent to which the consensus embraces the governments implementing the policy reforms is a matter of some controversy, which has been treated by John Toye, Paul Mosley and Jane Harrigan in their recent book. The term Washington Consensus was coined by Hans Singer precisely to emphasize the narrow base of the agreed programme.10 In what follows I shall show great restraint in not referring to the Washington Consensus as the WC though the appropriateness of the initials makes their use highly tempting. Several of the

Washington Consensus items refer to demand management; that is, restricting the growth of aggregate demand in the economy, the purpose of which would be to reduce inflation. These constitute the stabilisation aspect of the policy package. Structural adjustment is defined as supply side reforms. To the neoclassical economist structural adjustment means waving the magic wand of competition to generate economic efficiency. This would be achieved, in the consensus approach (1) through deregulating the capital market, both internally and with respect to the international economy, so that the interest rate is market determined and the national currency freely convertible (2) devaluation to make the exchange rate competitive, (3) elimination of trade restrictions, and (4) deregulation of domestic markets, especially the labour market. Capital market deregulation is supposed to generate an efficient allocation of capital and an optimal division between saving and consumption by households

15

Devaluation should shift resources from production of what are called non-traded commodities to exports and substitution for imports. Elimination of trade restrictions is predicted to sort out the chaff from the wheat in the economy, eliminating the inefficient producers and rewarding the efficient. Deregulation of internal markets will allow supply and demand to rule. In the labour market this will result in the reduction of unemployment through lower real wages and fewer restrictions on capitalists in the hiring and firing of workers. By getting the government out of the business of business, the Promethean invisible hand is unbound. Some of these measures taken alone seem reasonable; for example, marketing boards in some

African countries excessively taxed agricultural exports and exchange rates in many countries were allowed to appreciate real terms (as in the United Kingdom until a year ago). But each item must be viewed not on its own, but rather as a credible detail within a fantastic fairy tale; e.g., one does not challenge the possibility of inserting a pin cushion into the head of a scarecrow (Wizard of Oz), but granting that does not require one to conclude that it converts the scarecrow into an intellectual. Neoclassical stories share with fairy tales the use of the mundane and the microcredible to seduce the listener into embracing a mega-myth. The benign outcome of these measures is conditional upon the assumption that competition will increase and markets behave according to Walrus. Competition would certainly increase, but not in the manner of Walras. In the 1980s, the

liberalisation of financial markets in Latin America resulted in massive capital flight, until domestic interest rates rose above the levels in developed countries. A market determined interest rate meant determined by the US Federal Reserve System and the Bundesbank. Since base rates in the developed countries were at a historical high, the result was to increase dramatically the cost of credit in developing countries and depress investment.11 The argument that devaluations shift resources to more

efficient uses merely serves as a sugar coating for beggar thy neighbour policies in a demand-constrained world. Price cutting, which is the impact-effect of devaluation, may improve the ability to sell in a market. This is achieved in a short period of time by taking market shares from other producers. Recent work by Singer provides empirical evidence for a destructive and competitive devaluation process among Latin American countries in the 1980s.12 This is not to say that countries should never devalue; sometimes they should. But devaluation should not be encouraged as a

16

normal and recurrent tool of international competition, especially by multilateral agencies formally assigned the responsibility to seek the general welfare of nations. Recall that a motivation to create the IMF was in part to prevent a repeat of the competitive devaluations of the 1930s which contributed to the collapse of international trade. In the mid-1980s the World Bank advised the governments of developing countries that those which embarked upon trade liberalisation could look forward to a bright future no matter what other countries do. Here the fairy-tale masquerades as multilateral expertise. This startling conclusion (do not merely turn the other cheek, but lift your chin and step into the blow) requires two improbable assumptions on~ beginning: full employment and smooth mobility of resources (e.g., that a brewery can be converted on short notice to produce computer chips). The actual effect of trade liberalisation at best has been mixed. When trade liberalisation has resulted in bankrupt industries, unemployed workers, and destitute farmers in developing countries, the neoclassical argument comes full circle: those who point out that the fairy tale failed to make everyone live happily ever after are accused of being protectionists for blaspheming the free trade faith with an accusation of failure. It is also worth noting that the major governments pressing for free trade by developing countries hardly practice it themselves, though developing countries would be no better of if they did. The proposition that labour market deregulation will foster employment derives from Walrasian market clearing. On the basis of a fictional auctioneer, on the assumption that workers come each day to the market unemployed and to strike their bargains afresh, with the faith that all other markets simultaneously equilibrate, governments would abandon minimum wages, relax restrictions on hours of work, limit job security, and eliminate many controls on work-place safety. There is little theoretical reason to believe that such measures would have any positive impact on employment. There is considerable reason to believe that labour market deregulation results in falling wages, longer hours for those in work, and less safe working conditions; in other words, lower labour costs. In the inverted image provided by the neoclassical prism, paying workers less is good thing because it increases the competitiveness of production. The Never-Never Land of perfect competition appears in practice as its opposite: idle workers and unutilised capacity in place of full employment, currency 17

flight and depressed investment instead of efficient allocation of capital, and falling world prices for primary products through competitive devaluations instead of the fanciful gains from trade. The reality of competition is destabilisation, and regulation by the state should not be viewed as distorting, but rather the means by which the benefits of competition are realised and the costs minimised. Competition is inherent in capitalism and lends the system its progressive dynamic. In doing so, it generates uneven development among countries, inequality within countries, and instability of world markets. The neoclassical theory of competition has been designed to preclude all of these outcomes, and, when they occur, the theory explains them as the legacy of malevolent government interventions.

Conclusion
The theoretical justification for the liberalisation agenda is hardly convincing. Equally unimpressive have been the attempts of the advocates of competition to demonstrate empirically the benefits of the competitive state of bliss. The theory is weak; empirical estimates suggest trivial benefits accrue (work by Frances Stewart, Paul Mosley and John Toye has made important contributions to the empirical assessment of liberalisation).13 Why, then, do these arguments persist and prevail in policy debate? In part it is because neoclassical economics is rather like a computer virus of the mind. So counter to reality is the software that it takes several years to load it into the cranial PC, three years for the simplest version and up to seven or eight for the advanced PhD version. Its effect on the hardware goes far beyond consuming almost the entire memory capacity; like a virus it contaminates the alternative software in the brain; all thought processes tend to pass through its logical mode. Once suitably infected by the virus, the previously rational mind comes to accept that all unemployment is voluntary, that making workers easier to fire increases employment, and, the most basic philosophical leap of faith, that human existence can and should be expressed through the pleasure from consumption. Throughout the process of loading the

software, it is necessary repeatedly the test to neoclassical neophyte to ensure that there remains no proclivity to allow experience from the real world to feed back on the formulation of theory and that no non-hedonistic value enters the analysis.

18

The virus has the potential in the appropriate political climate, such as we have now, to produce powerful policy parables based upon the faith in the virtues of free markets. The deregulation of markets tends to benefit capital over labour and small producers. Neoclassical economics in its political form is the ideology which justifies this bias in benefits as synonymous with the general welfare. The call for more competition through deregulation, be it in developed or underdeveloped countries, is the call for liberation of capital from the constraints on its ability to expand, a point made eloquently by Marx in his famous reply to Prouhdon (The Poverty of Philosophy). Albeit in different terms than Marx, Keynes persuaded a generation of economists that the liberation of capital from the constraints of the state resulted in the unleashing of instability and the waste of resources; that the pre-condition for economic efficiency is not perfect competition, but the intervention of the state to bring full employment. Far from a creator of distortions that block economic inefficiency, the interventions of the state are manner by which the fairy tale might come true. This is the Keynesian vision of the mixed capitalist economy, in which the state suppresses the destructive aspects of competition, ensures full employment, and thus allows society to reap the benefits of capitalist dynamism. Fortunately, we have the recent book by Michael Stewart to re-acquaint us with Keynes vision. In providing the analysis to support this view of a socialised (but certainly not socialist) economy, Keynes initiated a Copernican revolution for mainstream economics (among Ricardians and Marxists the methodological break had been made earlier and more dramatically). Keynes sought to jerk the profession out of Cartesian idealism (the a priori method) and convert it to a method in which theory and reality enjoined interactively after the manner of the experimental sciences.14 The

defenders of the Ptolemaic geocentric model of the universe eventually, after one hundred years, were swept away by heliocentric theory of Copernicus, Galileo and others, with their empirical revolution consolidated by Newtons theory of gravity. However, after two hundred years the Ptolemaic harmony of Adam Smiths invisible hand still dominates economics. The profession dedicates itself to the production of increasingly arcane and improbable epicycles to reconcile empirical fact with theoretical fantasy in the economic equivalent of a geocentric universe of perfect and harmonious spheres of competition. This would represent a harmless and even charmingly eccentric pursuit were not it the case that millions of people throughout

19

the countries of the world find themselves the victims of policies based on a fairy tale that produces inequality, inefficiency, and conflict. In a Newtonian reconstruction of economics competition, not intervention, would be revealed as the process generating unemployment, inequality and inefficient use of resources. Neoclassical economics has turned causality on its head, such that the negative outcome of market forces are called market failures. This is, indeed, a pernicious term, suggesting that in the general case markets generate benign outcomes. Far from being failures, misallocation of resources, inequality and, for the developing world, poverty and uneven development, are the natural and systemic consequences of the process of unregulated competition.

Footnotes and References

1 [F]lexible prices, reflecting demand and supply in local and world markets, are the best way to signal to farmers what, how much, and when to produce. World Bank, Sub-Saharan Africa: From Crisis to Sustainable Growth (Washington: World Bank, 1989), p. 91. 2 Graham Bannock, R. F. Baxter and Evan Davis, Dictionary of Economics (London: Hutchinson, 1987), p. 125. 3 Pearce is more succinct in this definition of competitive markets: a market in which a very large number of small buyers and sellers trade independently, and as such no one trader can significantly influence price. David W. Pearce, The Macmillan Dictionary of Modern Economics (London: Macmillan, 1981), pp. 74-5. 4 S. E. Stiegler and Glyn Thomas (ed.), A Dictionary of Economics and Commerce (London Pan, 1976). 5 The contradiction between the neoclassical definition of competition and the forms of competition has been treated insightfully in I. A. Clifton, Competition and the Evolution of the Capitalist Mode of Production, Cambridge Journal of Economics (June 1977). 6 Many of the initial gains [from trade liberalisation] come from removing such price distortions, which discourage consumers from spending their income in a way which reflects the opportunities open to them in the international markets. Ultimately the producer gains will follow too, as resources are switched from, say, uneconomic European farms to products in which European countries have a true advantage. Samuel Brittan, Where Gatts $200bn really comes from Financial Times (4 October 1993), p. 12. 7 An exception is David Evans, Comparative Advantage and Growth Trade and Development in Theory and Practice (London. Harvester/Wheatsheaf 1989) 8 This example involves invalid simplifications: first, in a Walrasian market, farmers are price-takers, so it is against the rules for them to offer any prices; and, second, the argument is partial equilibrium, not general equilibrium. It is possible that notional excess supply of potatoes at a false trading price set might transubstantiate into a cleared market for potatoes at a higher price in the general equilibrium price set. 9 Williamson, J., What Washington Means by Policy Reform in J. Williamson (ed.), Latin American Adjustment: How Much has Happened? (Washington: Institute for International Economics, 1990) 10 This issue is treated m some detail m Paul Mosley Jane Harngan, and John Toye, Aid and Power: the World Bank and Policy-based Lending (London: Routledge 1991) 11 For a summary of the evidence on this point, see Paul Mosley and John Weeks, Has Africa Recovered? Africas Adjustment in the 1980s reconsidered, World Development (October 1993).

20

12 H. W. Singer, The relationship between debt pressures, adjustment policies and deterioration of terms of trade for developing countries (with special reference to Latin America) IDS Working Paper Series. No.59 (July 1989). 13 See, for example, Giovanni Andrea Cornia and Frances Stewart, The Fiscal System, Adjustment and the Poor, Centro Studi Luca dAngliano, Development Studies Working Papers, No. 29 (September 1990), and Mosley, Harrigan and Toye, 1991. 14 In this context a classic work is Victoria Chicks Macroeconomics after Keynes (London: Philip Allan, 1983).

21